filmov
tv
Lecture 4 Part 2: Nonlinear Root Finding, Optimization, and Adjoint Gradient Methods
Показать описание
MIT 18.S096 Matrix Calculus For Machine Learning And Beyond, IAP 2023
Instructors: Alan Edelman, Steven G. Johnson
Description: Nonlinear root finding by Newton’s method and optimization by gradient descent. “Adjoint” methods (reverse-mode/backpropagation) lets us find gradients efficiently for large-scale engineering optimization.
License: Creative Commons BY-NC-SA
Instructors: Alan Edelman, Steven G. Johnson
Description: Nonlinear root finding by Newton’s method and optimization by gradient descent. “Adjoint” methods (reverse-mode/backpropagation) lets us find gradients efficiently for large-scale engineering optimization.
License: Creative Commons BY-NC-SA